The hostname part of a DNS FQDN can allow for additional characters
other than specified in `man 7 hostname`.
This extends is_dnsaddr and the test issue #5657.
Also fixes a typo in a comment.
Changelog-Fixed: wireaddr: #5657 allow '_' underscore in hostname part of DNS FQDN
This was reported, but the channel was closed. So however we ended
up with a duplicate, we're no *worse* off than we were before migration?
Fixes: #5760
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
It's obviously the incorrect type, while our CI didn't catch it, Nicholas did:
```
plugins/fetchinvoice.c:1362:30: error: conversion from 'long long unsigned int' to 'size_t' {aka 'unsigned int'} changes value from '18446744073709551615' to '4294967295' [-Werror=overflow]
1362 | || tlv_span(wire, 1001, UINT64_MAX, NULL) != 0) {
```
Reported-by: @NicholasDorier
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Fixed: Build: arm32 compiler error in fetchinvoice, due to bad types on 32-bit platforms.
Fixes: #5776
This reverts commit 43b037ab0b.
Nicholas Dorier says BTC Payserver still wants this for another year
or so.
Changelog-Added: JSON-RPC: reverts requirement for "jsonrpc" "2.0" inside requests (still deprecated though, just for a while longer!)
The switch to logging enabled verbose output regardless of the option
flag. Here the functionality is restored.
Changelog-Fixed: reckless verbosity properly applied.
The read_json() call expects len_read to be the amount of *new*
data read. If we call this back without resetting, it will parse
this much random junk in the buffer.
Fixes: #5766
Changelog-Fixed: Plugin: `autoclean` could misperform or get killed due to lightningd's invalid handling of JSON batching.
We only incremented this if it was the wrong state.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Fixed: JSON-RPC: `autoclean-once` response `uncleaned` count is now correct.
We introduced a minor issue in #5757 that was causing the manpages to
be added every time we regenerate instead of replacing them. The
script was a bit unscrutable, and we do this block-replacement in
several places I thought it might be a good idea to have a dedicated
tool to do it.
This allows us to have simpler Makefiles whenever we update a
generated block inside another file. It's python, but does not have
dependencies :-)
Changelog-None
We may end up with a short read, that is not empty, when a message
gets fragmented during transport, so we need to loop until we either
reach the full size or we have an empty read indicating a dropped
connection.
Changelog-None
If we both support large channels, we can actually send giant HTLCs.
This, in turn, fixes pay (which relies on this field!).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Fixed: Plugins: `pay` now knows it can use locally-connected wumbo channels for large payments.
Fixes: #5250Fixes: #5417
This is a mistake that I introduced while I implemented the lightningd custom error for hsmd.
In particular, when the command line parser check
if the file is encrypted make an additional check regarding the existence of the file.
This can be a not useful check, but due to the delicate nature of the hsm file, it is better to check if exist and if it doesn't exist an informative line inside the log is emitted that notifies the user that the file does not exist.
This log may return useful while debugging disaster
(that can happen) to understand that there is something strange (eg. the user moves the hsm file somewhere else).
Fixes https://github.com/ElementsProject/lightning/issues/5719
Changelog-Fixed: lightningd: do not abort while parsing hsm pwd
Signed-off-by: Vincenzo Palazzo <vincenzopalazzodev@gmail.com>
This was found by tests/test_plugin.py::test_important_plugin and
was NOT a flake!
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-None: only just committed
This is the official root repository, and some tools (pip for example) doesn't like
pointing to repos that don't actually contain the commit that we point to. Funny
that Github actually shows non-existent commits in clones (bitcoin/bitcoin has been
impersonated before too).
We have a primary key that is spanning the `in_channel_id` and the
`in_htcl_id`. The latter gets set to NULL when the HTLC and channel
gets deleted, so we coalesce with a random large number that is
unlikely to collide for the primary key.
This is a minimal fix: we wait until all plugins reply from init before
continuing. Really large or busy nodes can have other things monopolize
lightningd, then the timer goes off and we blame the plugin (which has
responded, we just haven't read it yet!).
The real answer is to have some timeouts only advance when we're idle,
or have them low-priority so we only activate them when we're idle (this
doesn't apply to all timers: some are probably important!). But
this is a minimal fix for -rc3.
Fixes: https://github.com/ElementsProject/lightning/issues/5736
Changelog-Fixed: plugins: on large/slow nodes we could blame plugins for failing to answer init in time, when we were just slow.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
And internal changes don't get CHANGELOG entries, so removed
"- Added interactive transaction building routine ([#5287])".
We need to come up with some way of marking rust crate changes,
as it's not immediately obvious what "- cln-plugin" means?
Maybe we switch from JSON-RPC to Control as a more general
prefix?
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The autoclean plugin would assume we have a `resolved_time` which may
not be true for oldish nodes that predate our annotations.
Changelog-None Unreleased change
Reported-by: <@devastgh>
And document support for it.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Added: Plugins: `getmanfest` response can contain `nonnumericids` to indicate support for modern string-based JSON request ids.
Changelog-Deprecated: Plugins: numeric JSON request ids: modern ones will be strings (see doc/lightningd-rpc.7.md!)
The `optional` keyword was first removed, and then re-added again, and
then it was made into an experimental flag. Come on people, decide
already which one it is...
We had a quadratic check when processing a block, and this just
reduces it to be linear. Combined with the previous fix of adding txs
multiple times, this should speed up block processing considerably.
We were using a quadratic lookup to locate the transactions we
broadcast, so let's use an `htable` instead, allowing for constant
lookups during block processing.
The list of outgoing transactions was growing with every
rebroadcast. Now we just check whether it is already in the list and
don't add it anymore
Changelog-Fixed: ld: Reduce identification of own transactions to not slow down over time, reducing block processing time
Suggested-by: Matt Whitlock <@whitslack>
On testnet I noticed if we can't reach bitcoind for some reason, we'll
keep RBFing our penalty tx ("it didn't go in, RBF harder!!"). Makes
no sense to grossly exceed the amount needed for next block, so simply
cap penalty at 2x "estimatesmartfee 2 CONSERVATIVE".
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Ubuntu clang 15.0.2-1 complains:
```
wallet/wallet.c:280:6: error: variable 'i' set but not used
[-Werror,-Wunused-but-set-variable]
int i;
^
wallet/wallet.c:339:6: error: variable 'i' set but not used
[-Werror,-Wunused-but-set-variable]
int i;
^
wallet/wallet.c:4768:9: error: variable 'count' set but not used
[-Werror,-Wunused-but-set-variable]
size_t count;
^
```
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Ubuntu clang 15.0.2-1 complains:
```
wire/peer_exp_wiregen.c:257:14: error: variable 'i' set but not used [-Werror,-Wunused-but-set-variable]
for (size_t i = 0; *plen != 0; i++) {
^
wire/peer_exp_wiregen.c:1373:14: error: variable 'i' set but not used [-Werror,-Wunused-but-set-variable]
for (size_t i = 0; *plen != 0; i++) {
```
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
